nature microbiology
Article
https://doi.org/10.1038/s41564-022-01272-z
Comparative effectiveness of third doses
of mRNA-based COVID-19 vaccines in
US veterans
Received: 28 July 2022
Accepted: 17 October 2022
Published online: xx xx xxxx
Barbra A. Dickerman 1,2,9, Hanna Gerlovin 3,9 , Arin L. Madenci 1,2,4,
Michael J. Figueroa Muñiz 3,5, Jessica K. Wise3, Nimish Adhikari3,5,
Brian R. Ferolito3, Katherine E. Kurgansky3,6, David R. Gagnon 3,5, Kelly Cho3,7,
Juan P. Casas3,7 & Miguel A. Hernán 1,2,8
Check for updates
Vaccination against SARS-CoV-2 has been effective in reducing the burden
of severe disease and death from COVID-19. Third doses of mRNA-based
vaccines have provided a way to address waning immunity and broaden
protection against emerging SARS-CoV-2 variants. However, their
comparative effectiveness for a range of COVID-19 outcomes across diverse
populations is unknown. We emulated a target trial using electronic health
records of US veterans who received a third dose of either BNT162b2 or
mRNA-1273 vaccines between 20 October 2021 and 8 February 2022, during
a period that included Delta- and Omicron-variant waves. Eligible veterans
had previously completed an mRNA vaccine primary series. We matched
recipients of each vaccine in a 1:1 ratio according to recorded risk factors.
Each vaccine group included 65,196 persons. The excess number of events
over 16 weeks per 10,000 persons for BNT162b2 compared with mRNA1273 was 45.4 (95% CI: 19.4, 84.7) for documented infection, 3.7 (2.2, 14.1)
for symptomatic COVID-19, 10.6 (5.1, 19.7) for COVID-19 hospitalization,
2.0 (−3.1, 6.3) for COVID-19 intensive care unit admission and 0.2 (−2.2, 4.0)
for COVID-19 death. After emulating a second target trial of veterans who
received a third dose between 1 January and 1 March 2022, during a period
restricted to Omicron-variant predominance, the excess number of events
over 9 weeks per 10,000 persons for BNT162b2 compared with mRNA-1273
was 63.2 (95% CI: 15.2, 100.7) for documented infection. The 16-week risks of
COVID-19 outcomes were low after a third dose of mRNA-1273 or BNT162b2,
although risks were lower with mRNA-1273 than with BNT162b2, particularly
for documented infection.
Additional doses of messenger RNA (mRNA)-based vaccines are being
widely deployed to counter waning immunity and broaden protection against emerging variants of severe acute respiratory syndrome
coronavirus 2 (SARS-CoV-2)1,2. However, head-to-head comparisons
A full list of affiliations appears at the end of the paper.
Nature Microbiology
of the effectiveness of a third dose of different mRNA-based vaccines
have been lacking.
In a previous head-to-head comparison of breakthrough COVID-19
outcomes after the first dose of the BNT162b2 (Pfizer-BioNTech)
e-mail: hanna.gerlovin@va.gov
Article
https://doi.org/10.1038/s41564-022-01272-z
3,175,771 veterans ≥65 yr old, or veterans 18–64 yr old with high risk of severe COVID-19 between 20 October 2021
and 18 November 2021, or veterans ≥18 yr old between 19 November 2021 and 8 February 2022, with documented receipt
of the second dose of an mRNA COVID-19 vaccine at least 6 months earlier were identified in the VA database
1,170,681 received a third
dose of BNT162b2 or mRNA1273 before 8 February 2022
Had an interaction with the healthcare
system within 3 d before the vaccination
date (39,382)
Did not have a known residential address or
were in long-term care (28,735)
Were not determined to be a user of the VA
healthcare system in the past year (26,403)
Did not have recent data on body-mass index
or smoking (392,102)
Had an incomplete COVID-19 vaccination
record (256,429)
427,630 were evaluated for
inclusion in the study
170,748 received a
BNT162b2 third dose
256,882 received an
mRNA-1273 third dose
147,553 attended a VA
station that administered the
mRNA-1273 vaccine and
were eligible for subsequent
analyses
65,196 were included in the
matched cohort
214,728 attended a VA
station that administered the
BNT162b2 vaccine and were
eligible for subsequent
analyses
1:1
matching
65,196 were included in the
matched cohort
Fig. 1 | Selection of persons for the emulation of a target trial evaluating the comparative effectiveness of a third dose of BNT162b2 and mRNA-1273 vaccines.
Selection was conducted during a period spanning Delta- and Omicron-variant predominance (20 October 2021 to 8 February 2022).
and mRNA-1273 (Moderna) vaccines among 439,684 US veterans
aged ≥18 yr, we found a low risk of documented infection and severe
COVID-19 outcomes (for example, hospitalization and death) in a
period marked by SARS-CoV-2 Alpha-variant predominance in both
vaccine groups, but lower risk for the mRNA-1273 vaccine (with a similar
pattern for documented infection in a period marked by Delta-variant
predominance)3. It is unclear whether the same findings apply to third
doses and a range of COVID-19 outcomes in periods marked by other
SARS-CoV-2 variants, including Omicron.
A report from Spain estimated a third dose of mRNA-1273 to be 13%
more effective than a third dose of BNT162b2 for the prevention of documented SARS-CoV-2 infection during a period of Omicron-variant predominance4. In other reports, effectiveness of mRNA-based vaccines
against symptomatic COVID-19 was analysed separately for each third
dose compared with no vaccination5 (or with the primary vaccination
series6). These comparisons provide indirect evidence for comparative effectiveness but only if the comparators from each analysis had
Nature Microbiology
similar characteristics. Additional studies are needed to compare the
effectiveness of a third dose of these vaccines head-to-head for severe
COVID-19 outcomes to inform the choice of vaccine in coordinated
public health responses. Ideally, any comparative effectiveness study
should include racially diverse groups, evaluate potential differences
in effectiveness according to time since the completion of the primary
vaccination series and separately address time frames that include
predominance of different SARS-CoV-2 variants.
Here we analysed data from the national healthcare databases of
the Department of Veterans Affairs (VA), the largest integrated healthcare system in the US, to compare the effectiveness of a third dose of
either the BNT162b2 or the mRNA-1273 vaccine among US veterans who
had completed an mRNA vaccine primary series and received a third
dose between (1) 20 October 2021 and 8 February 2022 (a period spanning Delta- and Omicron-variant predominance) or (2) 1 January and 1
March 2022 (a period restricted to Omicron-variant predominance).
Recipients of each vaccine were matched in a 1:1 ratio according to their
Article
https://doi.org/10.1038/s41564-022-01272-z
Table 1 | Baseline characteristics of matched persons in
the target trial emulation evaluating the comparative
effectiveness of a third dose of BNT162b2 and mRNA1273 vaccines during a period spanning Delta- and
Omicron-variant predominance
Characteristic
Median age (IQR) (yr)
BNT162b2
recipients
(N = 65,196)
mRNA-1273
recipients
(N = 65,196)
70.0 (62.0, 74.0)
70.0 (62.0, 74.0)
Characteristic
BNT162b2
recipients
(N = 65,196)
mRNA-1273
recipients
(N = 65,196)
Heterologous
3,278 (5.0)
2,274 (3.5)
No. of SARS-CoV-2 tests performed in the past year, no. (%)
0
43,527 (66.8)
43,527 (66.8)
1
10,868 (16.7)
10,868 (16.7)
≥2
10,801 (16.6)
10,801 (16.6)
No. of primary care visits in the past 5 yr, no. (%)
Age group, no. (%)
18–39 yr
1,005 (1.5)
1,028 (1.6)
1–9
4,541 (7.0)
3,545 (5.4)
40–49 yr
2,247 (3.4)
2,217 (3.4)
10–19
21,014 (32.2)
18,301 (28.1)
20–29
18,356 (28.2)
18,833 (28.9)
21,285 (32.6)
24,517 (37.6)
50–59 yr
8,687 (13.3)
8,614 (13.2)
60–69 yr
20,029 (30.7)
20,092 (30.8)
≥30
70–79 yr
27,623 (42.4)
27,635 (42.4)
No. of influenza vaccinations in the past 5 yr, no. (%)
≥80 yr
5,605 (8.6)
5,610 (8.6)
0
9,240 (14.2)
7,667 (11.8)
Sex, no. (%)
1 or 2
11,974 (18.4)
10,120 (15.5)
Male
62,485 (95.8)
62,485 (95.8)
3 or 4
23,236 (35.6)
22,222 (34.1)
Female
2,711 (4.2)
2,711 (4.2)
≥5
20,746 (31.8)
25,187 (38.6)
Race, no. (%)
Black
15,365 (23.6)
15,365 (23.6)
Other
705 (1.1)
705 (1.1)
Unknown
1,013 (1.6)
1,013 (1.6)
White
48,113 (73.8)
48,113 (73.8)
Hispanic
4,224 (6.5)
5,878 (9.0)
Not Hispanic
59,446 (91.2)
57,848 (88.7)
Unknown
1,526 (2.3)
1,470 (2.3)
Urban residence, no. (%)
50,919 (78.1)
50,919 (78.1)
Current
21,926 (33.6)
23,013 (35.3)
Former
21,517 (33.0)
20,234 (31.0)
Never
21,753 (33.4)
21,949 (33.7)
Chronic lung diseasea
10,715 (16.4)
11,913 (18.3)
Cardiovascular diseaseb
18,341 (28.1)
18,589 (28.5)
Hypertension
42,726 (65.5)
44,320 (68.0)
Diabetes
23,053 (35.4)
25,414 (39.0)
Chronic kidney disease
6,163 (9.5)
6,889 (10.6)
Chronic liver disease
2,540 (3.9)
2,241 (3.4)
Cancerc
8,877 (13.6)
8,936 (13.7)
Immunocompromised stated
3,707 (5.7)
4,213 (6.5)
Obesitye
30,091 (46.2)
30,714 (47.1)
Dementia
1,165 (1.8)
1,089 (1.7)
Substance use disorder
5,164 (7.9)
5,211 (8.0)
Ethnicity, no. (%)
Smoking status, no. (%)
Coexisting conditions, no. (%)
Months since completion of mRNA COVID-19 vaccine primary series, no. (%)
6–7
20,259 (31.1)
20,071 (30.8)
8
25,951 (39.8)
26,077 (40.0)
≥9
18,986 (29.1)
19,048 (29.2)
Vaccine type for mRNA COVID-19 vaccine primary series (compared with third
dose type), no. (%)
Homologous
Nature Microbiology
61,918 (95.0)
62,922 (96.5)
Persons included in this target trial emulation received a third dose of BNT162b2 or mRNA1273 between 20 October 2021 and 8 February 2022.
a
Chronic lung disease included asthma, bronchitis and chronic obstructive pulmonary
disease. bCardiovascular disease included acute myocardial infarction, cardiomyopathy,
cerebrovascular disease, coronary heart disease, heart failure and peripheral vascular
disease. cNot included here are non-melanoma skin cancer, benign neoplasms, cancers
in situ and neoplasms of uncertain behaviour. dImmunocompromised state included
human immunodeficiency virus infection, organ or tissue transplant, bone marrow biopsy,
or use of any of the following medications (prescribed ≥2 times over the past year):
systemic glucocorticoids, anti-inflammatory or anti-rheumatic agents in combination with
glucocorticoids and immunosuppressants. eObesity was defined as a body-mass index (the
weight in kilograms divided by the square of the height in metres) of 30 or greater. IQR,
interquartile range. Percentages may not total to 100% because of rounding.
risk factors. Comparative effectiveness was estimated over 16 weeks in
the combined Delta-Omicron period and over 9 weeks in the Omicron
period for five COVID-19 outcomes: documented SARS-CoV-2 infection,
symptomatic COVID-19 and COVID-19-related hospitalization, admission to an intensive care unit (ICU) and death.
Results
Primary analyses during a period spanning SARS-CoV-2 Deltaand Omicron-variant predominance
Study population and follow-up. Among 147,553 eligible recipients of
a BNT162b2 third dose and 214,728 eligible recipients of an mRNA-1273
third dose, 65,196 BNT162b2 recipients were matched to 65,196 mRNA1273 recipients (Fig. 1). The baseline characteristics of this matched
population (Table 1) were similar to those of the eligible population
(Supplementary Table 1). The median age was 70 yr (interquartile
range, 62–74 yr), 96% of individuals were male and 24% were Black.
The two vaccine groups had similar distributions of demographics,
coexisting conditions and markers of healthcare utilization (see also
Extended Data Fig. 1).
The median follow-up was 77 d (interquartile range, 61 to 97 d).
Over a 16-week-follow-up, 2,994 SARS-CoV-2 infections were documented, of which 200 were detected as symptomatic COVID-19 within
the VA healthcare system, 194 required hospitalization, 52 required ICU
admission and 22 resulted in death.
Comparative effectiveness. Over a 16-week period spanning Deltaand Omicron-variant predominance, the estimated risk of documented
infection was 353.9 (95% CI: 326.7, 373.2) events per 10,000 persons for
the BNT162b2 third dose and 308.5 (95% CI: 276.8, 320.9) events per
10,000 persons for the mRNA-1273 third dose (Fig. 2). As expected, we
found a nearly identical risk pattern in the two vaccine groups in the
Article
https://doi.org/10.1038/s41564-022-01272-z
a
b
Documented SARS−CoV−2 infection
Symptomatic COVID−19
0.5%
BNT162b2
3.0%
Cumulative incidence (%)
Cumulative incidence (%)
4.0%
mRNA−1273
2.0%
1.0%
0
BNT162b2
0.4%
mRNA−1273
0.3%
0.2%
0.1%
0
0
14
28
42
56
70
84
98
112
0
14
28
42
Time (d)
c
d
COVID−19 hospitalization
BNT162b2
0.4%
Cumulative incidence (%)
Cumulative incidence (%)
70
84
98
112
70
84
98
112
COVID−19 ICU admission
0.15%
0.5%
mRNA−1273
0.3%
0.2%
0.1%
0
BNT162b2
mRNA−1273
0.10%
0.05%
0
0
14
28
42
56
70
84
98
112
Time (d)
e
56
Time (d)
0
14
28
42
56
Time (d)
COVID−19 death
Cumulative incidence (%)
0.15%
BNT162b2
mRNA−1273
0.10%
0.05%
0
0
14
28
42
56
70
84
98
112
Time (d)
Fig. 2 | Cumulative incidence of Covid-19 outcomes during a period spanning
Delta- and Omicron-variant predominance (20 October 2021 to 15 February
2022). a–e, Documented SARS-CoV-2 infection (a), symptomatic COVID-19
(b), COVID-19 hospitalization (c), COVID-19 ICU admission (d) and COVID-19
deaths (e) during a period spanning Delta- and Omicron-variant predominance
(20 October 2021 to 15 February 2022). Solid blue line represents the risk curve
for BNT162b2, dashed orange line represents the risk curve for mRNA-1273 and
shaded areas represent pointwise 95% confidence intervals.
evaluations of two negative outcome controls7: symptomatic COVID-19
during the first 7 d after the third vaccine dose (Extended Data Fig. 2)
and death from causes other than COVID-19 during follow-up (Extended
Data Fig. 3).
The estimated 16-week risk ratios (95% CI) for recipients of a third
dose of BNT162b2 as compared with mRNA-1273 were 1.15 (1.06, 1.30)
for documented SARS-CoV-2 infection, 1.21 (1.12, 2.14) for symptomatic
COVID-19, 1.64 (1.27, 2.79) for COVID-19 hospitalization, 1.37 (0.67, 3.14)
for COVID-19 ICU admission and 1.08 (0.46, 6.39) for COVID-19 death
(Table 2). The estimated risk differences (BNT162b2 minus mRNA-1273),
expressed as events over 16 weeks per 10,000 persons (95% CI), were
45.4 (19.4, 84.7) for documented SARS-CoV-2 infection, 3.7 (2.2, 14.1)
for symptomatic COVID-19, 10.6 (5.1, 19.7) for COVID-19 hospitalization, 2.0 (−3.1, 6.3) for COVID-19 ICU admission and 0.2 (−2.2, 4.0) for
COVID-19 death.
Estimates were similar across subgroups defined at baseline
according to age, race, months since completion of COVID-19 vaccine
primary series (Table 3) and by vaccine type of primary series (risk
Nature Microbiology
Article
https://doi.org/10.1038/s41564-022-01272-z
Table 2 | Estimated comparative effectiveness of a third dose of BNT162b2 and mRNA-1273 vaccines during a period
spanning Delta- and Omicron-variant predominance (20 October 2021 to 15 February 2022)
COVID-19 outcomes
No. of events
BNT162b2
16-week risk no. of events/10,000 persons
(95% CI)
mRNA-1273 BNT162b2
mRNA-1273
Risk difference no. of
events/10,000 persons (95% CI)
Risk ratio (95% CI)
BNT162b2 vs mRNA-1273
Documented infection
1,640
1,354
353.9 (326.7, 373.2)
308.5 (276.8, 320.9)
45.4 (19.4, 84.7)
1.15 (1.06, 1.30)
Symptomatic COVID-19
114
86
21.6 (18.3, 27.9)
17.9 (11.2, 18.3)
3.7 (2.2, 14.1)
1.21 (1.12, 2.14)
a
Hospitalization
126
68
27.2 (21.7, 32.7)
16.6 (10.2, 20.2)
10.6 (5.1, 19.7)
1.64 (1.27, 2.79)
ICU admission
33
19
7.5 (4.8, 10.5)
5.5 (2.6, 10.0)
2.0 (−3.1, 6.3)
1.37 (0.67, 3.14)
Death
13
9
2.9 (1.6, 5.5)
2.7 (0.6, 4.9)
0.2 (−2.2, 4.0)
1.08 (0.46, 6.39)
Recipients of a third dose of the BNT162b2 vaccine were matched in a 1:1 ratio to recipients of a third dose of the mRNA-1273 vaccine according to the following variables: calendar date of third
dose, calendar month of second mRNA vaccine dose, age, sex, race, urbanicity of residence, geographic location coded as categories of Veterans Integrated Services Network and number of
SARS-CoV-2 tests performed in the past year.
a
The number of tests for SARS-CoV-2 infection recorded during the study follow-up was 14,595 for the BNT162b2 group and 14,502 for the mRNA-1273 group. CI, confidence interval.
ratio for documented infection comparing a third dose of BNT162b2 vs
mRNA-1273, 1.19 (0.98, 2.65) among recipients of a BNT162b2 primary
series, 1.04 (0.84, 2.32) among recipients of an mRNA-1273 primary
series). Estimates were also similar in sensitivity analyses that excluded
eligible individuals with previously documented SARS-CoV-2 infection
(6.3%), those who did not receive a homologous COVID-19 vaccine primary series (3.3%) and those whose third dose could not be identified
as a booster dose (17.2%) (data not shown).
Secondary analyses during a period of SARS-CoV-2
Omicron-variant predominance
Study population and follow-up. Among 25,557 eligible recipients of
a BNT162b2 third dose and 36,809 eligible recipients of an mRNA-1273
third dose, 7,894 BNT162b2 recipients were matched to 7,894 mRNA1273 recipients. Compared with the eligible population, the matched
population was generally similar with respect to baseline demographic
and clinical characteristics but included a higher percentage of men,
White individuals, individuals with an urban residence and individuals
who had received no SARS-CoV-2 tests in the past year (Supplementary
Table 1). Baseline characteristics of the matched persons are shown in
Supplementary Table 2. All variables were well-balanced between the
two vaccine groups (Extended Data Fig. 1).
As compared with the matched population during a period spanning Delta- and Omicron-variant predominance, the matched population during a period of Omicron-variant predominance was, on average,
younger and included a higher percentage of individuals who had
received no SARS-CoV-2 tests in the previous year and no influenza
vaccinations in the previous 5 yr at a VA facility (Supplementary Table 1).
The median follow-up was 54 d (interquartile range, 40–60 d).
Over a 9-week follow-up, 214 SARS-CoV-2 infections were documented.
Comparative effectiveness. Over a 9-week period of Omicron-variant
predominance, the estimated risk of documented SARS-CoV-2 infection
was also higher with a third dose of the BNT162b2 vaccine than with a
third dose of the mRNA-1273 vaccine; the estimated risk ratio was 1.57
(1.12, 2.10) and the estimated risk difference, expressed as events over
9 weeks per 10,000 persons, was 63.2 (15.2, 100.7) (see also Fig. 3).
Discussion
We quantified the comparative effectiveness of a third dose of the
BNT162b2 and mRNA-1273 vaccines for the prevention of COVID-19
outcomes in the largest integrated healthcare system in the United
States. Although the risks of all five measured COVID-19 outcomes
(documented infection, symptomatic COVID-19 and COVID-19-related
hospitalization, ICU admission and death) over 16 weeks were low for
both vaccines during a period spanning Delta- and Omicron-variant
predominance (risks <4% for documented infection and <0.03%
for death, within each vaccine group), recipients of a third dose of
Nature Microbiology
BNT162b2 had an excess per 10,000 persons of 45 documented infections and 11 hospitalizations compared with recipients of a third dose
of mRNA-1273. We also found a higher risk of documented infection
among recipients of a third dose of BNT162b2 compared with mRNA1273 over 9 weeks of follow-up during a period of Omicron-variant
predominance, although this estimate was less precise due to a smaller
number of eligible persons.
Few head-to-head comparisons of a third dose of mRNA vaccines
are available. A report from Spain estimated mRNA-1273 boosters to
be 13% more effective than BNT162b2 boosters for the prevention of
documented SARS-CoV-2 infection during a period of Omicron-variant
predominance, but more severe outcomes could not be evaluated4. In
two other reports, effectiveness against symptomatic COVID-19 was
evaluated separately for each mRNA booster either compared with no
vaccination (in two test-negative case-control studies in England5) or
compared with the primary series (in two matched cohort studies in
Qatar6). In each report, an indirect comparison of these results requires
that the controls from each study had similar characteristics, and more
severe outcomes could not be evaluated for both mRNA booster groups.
The differential effectiveness we report in our study might be
explained by the higher mRNA content of mRNA-1273 (50 μg for booster
doses, 100 μg for third primary doses) compared with of BNT162b2
(30 μg for booster and third primary doses). Our primary analysis
considered any third dose of these vaccines, of which 83% were distinguished as booster doses, and results were similar in sensitivity analyses
that were restricted to these.
The strengths of our study are that first, the VA healthcare databases contain rich data on demographics and medical history, which
allowed us to carefully match recipients of each vaccine type according
to key confounders. Second, the databases contain detailed information on laboratory test results and healthcare encounters, which
allowed us to capture outcomes related to COVID-19 in both outpatient
and inpatient settings. Third, the large size of the study population
allowed us to evaluate less common COVID-19 outcomes (hospitalizations, ICU admissions, deaths). Fourth, the demographic composition of the US veteran population allowed us to provide evidence for
a diverse cohort (24% Black, 8% Hispanic) and to conduct subgroup
analyses among older persons (≥70 yr of age) and Black persons.
Our study also has several potential limitations. First, as in any
observational analysis of comparative effectiveness, the vaccine
groups could differ with respect to risk factors for the outcomes.
However, after rigorously matching recipients of each vaccine type,
the two groups had similar demographics, comorbidities, time since
completion of the primary vaccination series, number of SARS-CoV-2
tests in the previous year and markers of healthcare utilization (for
example, number of primary care visits and of influenza vaccinations
in the previous 5 yr). Further, much less confounding is expected when
comparing recipients of different third doses of vaccines than when
Article
https://doi.org/10.1038/s41564-022-01272-z
Table 3 | Estimated comparative effectiveness of a third dose of BNT162b2 and mRNA-1273 vaccines in subgroups defined
by baseline characteristics during a period spanning Delta- and Omicron-variant predominance (20 October 2021 to
15 February 2022)
No. of eventsa
16-week no. of events/10,000 persons (95% CI)
Risk difference no of
events/10,000 persons (95% CI)
Risk ratio (95% CI)
BNT162b2
mRNA-1273
BNT162b2
mRNA-1273
BNT162b2 vs mRNA-1273
Documented infection
888
723
409.1 (364.9, 441.5)
340.4 (294.4, 361.9)
68.8 (26.4, 122.1)
1.20 (1.07, 1.39)
Symptomatic COVID-19
56
40
21.0 (15.3, 26.8)
17.5 (9.6, 24.2)
3.5 (−5.0, 13.3)
1.20 (0.77, 2.21)
Hospitalization
67
24
30.5 (17.5, 33.1)
13.4 (6.8, 18.2)
17.1 (3.7, 22.7)
2.28 (1.23, 3.88)
ICU admission
19
6
9.5 (3.6, 13.3)
4.4 (0.4, 7.6)
5.1 (−1.1, 10.8)
2.16 (0.81, 15.00)
Death
–
–
–
–
–
–
Documented infection
716
617
301.8 (275.6, 332.9)
277.3 (245.1, 298.8)
24.5 (−9.0, 75.9)
1.09 (0.97, 1.31)
Symptomatic COVID-19
57
36
20.8 (16.2, 29.3)
13.9 (8.8, 18.3)
6.9 (1.1, 17.1)
1.50 (1.07, 2.76)
Hospitalization
63
34
26.5 (20.0, 34.9)
15.4 (9.7, 22.6)
11.1 (1.5, 21.0)
1.72 (1.08, 2.98)
ICU admission
17
13
7.4 (4.0, 11.4)
6.7 (2.3, 12.6)
0.8 (−6.2, 6.4)
1.11 (0.42, 3.31)
Death
12
8
5.0 (1.5, 7.4)
4.1 (1.0, 7.2)
0.9 (−4.0, 5.0)
1.21 (0.35, 5.37)
Documented infection
1,197
919
355.2 (315.0, 364.6)
287.8 (266.8, 314.1)
67.4 (13.4, 85.7)
1.23 (1.04, 1.31)
Symptomatic COVID-19
100
44
26.7 (17.9, 29.2)
11.3 (9.2, 18.3)
15.4 (2.5, 17.4)
2.37 (1.16, 2.59)
Hospitalization
99
45
28.0 (21.3, 33.7)
14.5 (9.4, 21.4)
13.4 (3.4, 20.8)
1.92 (1.17, 3.07)
ICU admission
26
15
7.4 (4.0, 10.2)
5.8 (2.3, 10.9)
1.7 (−5.3, 6.0)
1.29 (0.52, 3.14)
Death
14
9
4.4 (1.5, 6.2)
3.3 (0.6, 5.7)
1.1 (−2.8, 4.0)
1.35 (0.45, 6.34)
Documented infection
390
352
359.1 (317.4, 411.1)
318.4 (274.8, 353.0)
40.7 (−10.7, 115.0)
1.13 (0.97, 1.39)
Symptomatic COVID-19
26
22
19.7 (11.7, 28.1)
19.1 (10.1, 26.9)
0.6 (−10.4, 14.4)
1.03 (0.59, 2.37)
Hospitalization
29
18
29.9 (16.0, 43.4)
18.4 (7.4, 27.0)
11.5 (−3.7, 29.1)
1.63 (0.85, 4.41)
ICU admission
–
–
–
–
–
–
Death
–
–
–
–
–
–
Age <70 yr
Age ≥70 yr
Race, White
Race, Black
Months since completion of COVID-19 vaccine primary series, 6–7
Documented infection
508
389
330.5 (290.7,
358.3)
254.8 (227.6, 286.8)
75.7 (20.8, 108.3)
1.30 (1.08, 1.45)
Symptomatic COVID-19
35
19
20.2 (10.7, 23.4)
12.2 (6.8, 19.6)
8.0 (−4.6, 12.6)
1.66 (0.71, 2.75)
Hospitalization
34
12
23.2 (15.3, 31.3)
9.3 (5.2, 20.1)
13.9 (−0.2, 21.5)
2.49 (0.99, 4.33)
ICU admission
–
–
–
–
–
–
Death
–
–
–
–
–
–
Months since completion of COVID-19 vaccine primary series, 8
Documented infection
587
519
341.3 (306.1, 371.8)
320.5 (269.7, 347.0)
20.8 (−16.2, 86.3)
1.06 (0.95, 1.31)
Symptomatic COVID-19
46
25
23.1 (17.0, 33.8)
12.9 (8.4, 19.4)
10.2 (1.3, 21.4)
1.79 (1.07, 3.16)
Hospitalization
43
22
25.2 (17.1, 34.2)
13.4 (6.1, 18.0)
11.8 (3.2, 24.8)
1.88 (1.21, 4.54)
ICU admission
8
6
4.7 (2.1, 11.0)
3.9 (0.6, 7.0)
0.7 (−2.7, 8.0)
1.19 (0.48, 10.32)
Death
–
–
–
–
–
–
Months since completion of COVID-19 vaccine primary series, ≥9
Documented infection
395
315
305.9 (259.6,
334.6)
317.1 (233.8, 359.7)
−11.1 (−72.1, 78.0)
0.96 (0.79, 1.33)
Symptomatic COVID-19
32
20
23.9 (13.7, 31.7)
18.0 (9.0, 26.8)
5.9 (−7.4, 18.2)
1.33 (0.68, 2.55)
Hospitalization
43
21
34.9 (19.1, 43.6)
37.2 (11.9, 83.7)
−2.3 (−60.2, 24.1)
0.94 (0.30, 2.82)
ICU admission
15
6
11.3 (4.6, 16.2)
24.6 (1.2, 65.5)
−13.3 (−59.4, 12.4)
0.46 (0.09, 10.92)
Death
–
–
–
–
–
–
Recipients of a third dose of the BNT162b2 vaccine were matched in a 1:1 ratio to recipients of a third dose of the mRNA-1273 vaccine according to the following variables: calendar date of third
dose, calendar month of second mRNA vaccine dose, age, sex, race (except for subgroup analyses restricted to race), urbanicity of residence, geographic location coded as categories of
Veterans Integrated Services Network and number of SARS-CoV-2 tests performed in the past year. Estimates were calculated only in analyses in which there were more than 5 outcome events
in each vaccine group under comparison.
a
The sum of events across subgroups may not equal the sum of events in the overall population because the entire analysis (including matching) was repeated after stratification of the
population according to baseline characteristics.
Nature Microbiology
Article
https://doi.org/10.1038/s41564-022-01272-z
Omicron-variant predominance. Further evaluation of the comparative
effectiveness and safety of additional doses of these vaccines is needed.
Documented SARS−CoV−2 infection
3.0%
Methods
Cumulative incidence (%)
BNT162b2
Specification of the target trials
mRNA−1273
2.0%
1.0%
0
0
7
14
21
28
35
42
49
56
63
Time (d)
Fig. 3 | Cumulative incidence of documented SARS-CoV-2 infection during a
period of Omicron-variant predominance (1 January to 8 March 2022). Solid
blue line represents the risk curve for BNT162b2, dashed orange line represents
the risk curve for mRNA-1273 and shaded areas represent pointwise 95%
confidence intervals.
comparing recipients of a third dose with, for example, unvaccinated
persons. In addition, our two analyses involving negative-outcome
controls suggested little confounding.
Second, the possibility of outcome misclassification cannot be
ruled out if veterans sought testing or care outside the VA healthcare
system or if testing was differential between groups. However, our use
of the VA COVID-19 National Surveillance Tool allowed us to integrate
laboratory data with clinical notes to capture infections documented
inside and outside the system, and the SARS-CoV-2 testing frequencies over follow-up were similar in both groups (Table 2). Further, we
selected regular VA users who had a known residential address and
who had routinely collected information (body-mass index, smoking
status) to improve outcome ascertainment. We expect any residual
incomplete ascertainment to be non-differential between the vaccination groups under comparison, although absolute risks may be slightly
biased downwards. Finally, our study population was mostly made up
of men (96%) and older persons (95% were >50 yr old), which may limit
the generalizability of our findings.
Given the high effectiveness of a third dose of both the BNT162b2
and mRNA-1273 vaccines, either vaccine is strongly recommended to
any individual. This study provides evidence of clear and comparable
benefits of these vaccines for the most severe outcomes (the difference in estimated 16-week risk of death between the two groups was
two-thousandths of one percent). While the differences in estimated
risk for less severe outcomes between the two groups were small on
the absolute scale, they may be meaningful when considering the
population scale at which these vaccines are deployed. However, decisions about vaccination campaigns are complex and must incorporate
considerations that extend beyond the scope of the present study.
Therefore, our findings need to be combined with those from previous
studies to inform future decisions about choice of vaccine in coordinated public health responses.
In summary, although the absolute risks of breakthrough
SARS-CoV-2 infection and more severe COVID-19 outcomes were low
regardless of the third mRNA-based COVID-19 dose received, this study
involving a nationwide cohort of US veterans provides evidence of a
lower 16-week risk of COVID-19-related outcomes among recipients
of a third dose of the mRNA-1273 vaccine compared with recipients
of a third dose of the BNT162b2 vaccine, particularly for documented
infection. This pattern was consistent across a period spanning
Delta- and Omicron-variant predominance and a period restricted to
Nature Microbiology
We designed an observational analysis to emulate a hypothetical pragmatic trial of a third dose of BNT162b2 compared with mRNA-1273 and
risks of COVID-19 outcomes in the VA healthcare system. Supplementary Table 3 summarizes the key protocol components.
Eligibility criteria included veteran status; age ≥65 yr or 18–64 yr
with high risk of severe COVID-19 between 20 October and 18 November
2021 (based on the presence of at least one coexisting condition listed
in Table 1)8 or ≥18 yr between 19 November 2021 and 8 February 2022
(based on national guidelines for third dose deployment)9,10; receipt of
the second dose of an mRNA vaccine primary series at least 6 months
earlier (based on the same guidelines); known residential address
outside of a long-term care facility; and known smoking status and
known body-mass index within the previous year. Individuals had to
have used the VA healthcare system during the previous year (defined
as receiving care at a station eligible to administer the vaccines under
study and having at least one primary care visit) but not within the
previous 3 d (which may indicate the start of symptomatic disease).
The interventions of interest were a third dose of either the
BNT162b2 or the mRNA-1273 vaccine. To ensure balance of important
characteristics across groups, eligible veterans in the target trial would
be randomly assigned to one of these two vaccines within strata defined
by calendar date of the third dose (5 d bins), calendar month of the
second mRNA vaccine dose, age (5 yr bins), sex (male, female), race
(White, Black, other, unknown), urbanicity of residence (urban, not
urban), geographic location (coded as 19 categories of Veterans Integrated Services Network) and number of SARS-CoV-2 tests performed
in the past 12 months (0, 1, ≥2).
The five outcomes of interest were documented SARS-CoV-2 infection, documented symptomatic COVID-19 and COVID-19-related hospital admission, ICU admission and death. For each eligible individual,
follow-up started on the day the third dose of vaccine was received
(baseline) and ended on the day of the outcome of interest, death, 112 d
(16 weeks) after baseline, or the end of the study period (15 February
2022), whichever happened first.
Our target trial evaluates comparative effectiveness of a third
dose of the vaccines during a period spanning times when SARS-CoV-2
Delta- and Omicron-variants were circulating. The Delta variant had
decreased to a share of 26% of circulating variants in the United States
as of 25 December 2021, as it was rapidly displaced by the Omicron
variant, which rose to a 100% share as of 12 February 202211. To evaluate the comparative effectiveness during a period of Omicron-variant
predominance, we considered a second target trial that was identical to
the first trial except that the recruitment period was 1 January to 1 March
2022, and the only outcome of interest was documented SARS-CoV-2
infection because the period was too short to accrue a sufficient number of rarer, more severe outcomes.
Emulation of the target trials
We emulated both target trials using the VA healthcare databases3.
Vaccination was identified using records from the Corporate Data
Warehouse and the VA COVID-19 Shared Data Resource. SARS-CoV-2
infections were identified using the VA COVID-19 National Surveillance
Tool3,12, which integrates data on laboratory tests with natural language
processing of clinical notes to capture diagnoses documented inside
and outside the VA healthcare system. Symptomatic COVID-19 was
defined as ≥1 of the following symptoms documented within 4 d of
SARS-CoV-2 infection: fever, chills, cough, shortness of breath or difficulty breathing, sore throat, loss of taste or smell, headache, myalgia/
muscle pain, diarrhoea and vomiting. COVID-19 hospitalization was
defined as a hospitalization within 21 d after documented SARS-CoV-2
Article
infection, COVID-19 ICU admission was defined as an ICU admission
during a COVID-19 hospitalization, and COVID-19 death was defined
as a death within 30 d after documented SARS-CoV-2 infection. Supplementary Table 4 provides detailed information on all study variables
and their ascertainment.
We attempted to mimic the stratified randomization of the target
trial by matching persons who received a third dose of BNT162b2 and
of mRNA-1273 on the basis of the calendar date of the third dose, calendar month of the second dose, age, sex, race, urbanicity of residence,
geographic location, and number of SARS-CoV-2 tests performed in
the past 12 months, using the same matching algorithm described in
our previous study3.
To explore the possibility of residual confounding (for example,
by underlying health status or healthcare-seeking behaviour), we
incorporated two negative outcome controls7. First, we evaluated the
risk of symptomatic COVID-19 in the first 7 d after the third vaccine
dose. Second, we evaluated the risk of death from causes other than
COVID-19 during the follow-up period.
Statistical analyses
Covariate balance after matching was evaluated by plotting the mean
differences between variable values (standardized for continuous
variables) for the vaccination groups, with a difference of 0.1 or less
considered to be acceptable13.
Cumulative incidence (risk) curves for the vaccine groups were estimated using the Kaplan-Meier estimator14. We then calculated 16-week
risks of each outcome and compared them between the vaccine groups
via differences and ratios. We conducted subgroup analyses by age (<70
or ≥70 yr), race (Black or White), time since completion of the COVID-19
vaccine primary series (6–7, 8, or ≥9 months) and vaccine type of the
primary series (BNT162b2 or mRNA-1273). We conducted sensitivity
analyses that excluded eligible individuals who (1) had previously documented SARS-CoV-2 infection, (2) did not receive a homologous third
dose compared with the COVID-19 vaccine primary series and (3) had a
third dose that could not be identified as a booster dose on the basis of
procedural codes available in the ‘Inpatient’ and ‘Outpatient’ domains,
to compare recipients of vaccines at known doses, as the dose of mRNA1273 differs for booster doses (50 μg) vs third primary doses (100 μg).
We used a nonparametric bootstrapping procedure (including
both matching and subsequent analyses) with 500 iterations to calculate percentile-based 95% confidence intervals for all estimates.
Analyses were performed using R software version 3.6.0 (R Foundation for Statistical Computing) and SAS Enterprise Guide version
8.2 (SAS Institute).
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The data that support the findings of this study are available from
the VA. VA data are made freely available to researchers (behind the
VA firewall) with an approved VA study protocol. More information is
available at https://www.virec.research.va.gov or by contacting the VA
Information Resource Center (VIReC) at VIReC@va.gov. Source data
are provided with this paper.
Code availability
Access to the computer code used in this research is available by request
to the corresponding author.
References
1.
Garcia-Beltran, W. F. et al. mRNA-based COVID-19 vaccine
boosters induce neutralizing immunity against SARS-CoV-2
Omicron variant. Cell 185, 457–466.e4 (2022).
Nature Microbiology
https://doi.org/10.1038/s41564-022-01272-z
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Gruell, H. et al. mRNA booster immunization elicits potent
neutralizing serum activity against the SARS-CoV-2 Omicron
variant. Nat. Med. 28, 477–480 (2022).
Dickerman, B. A. et al. Comparative effectiveness of BNT162b2
and mRNA-1273 vaccines in U.S. veterans. N. Engl. J. Med. 386,
105–115 (2022).
Monge, S. et al. Effectiveness of mRNA vaccine boosters
against infection with the SARS-CoV-2 omicron (B.1.1.529) variant
in Spain: a nationwide cohort study. Lancet Infect. Dis. 22,
1313–1320 (2022).
Andrews, N. et al. Covid-19 vaccine effectiveness against
the Omicron (B.1.1.529) variant. N. Engl. J. Med. 386,
1532–1546 (2022).
Abu-Raddad, L. J. et al. Effect of mRNA vaccine boosters against
SARS-CoV-2 Omicron infection in Qatar. N. Engl. J. Med. 386,
1804–1816 (2022).
Lipsitch, M., Tchetgen Tchetgen, E. & Cohen, T.
Negative controls: a tool for detecting confounding
and bias in observational studies. Epidemiology 21,
383–388 (2010).
Underlying Medical Conditions Associated with Higher Risk
for Severe COVID-19: Information for Healthcare Professionals
(Centers for Disease Control and Prevention, accessed
3 March 2022).
FDA Authorizes Booster Dose of Pfizer-BioNTech COVID-19
Vaccine for Certain Populations [Press release] (US FDA,
22 September 2021).
Coronavirus (COVID-19) Update: FDA Takes Additional Actions on
the Use of a Booster Dose for COVID-19 Vaccines [Press release]
(US FDA, 20 October 2021).
COVID Data Tracker. Variant Proportions https://covid.cdc.gov/
covid-data-tracker/#variant-proportions (Centers for Disease
Control and Prevention, accessed 3 March 2022).
Chapman, A. et al. A natural language processing system for
national COVID-19 surveillance in the US Department of Veterans
Affairs. In Proc. 1st Workshop on NLP for COVID-19 at ACL 2020
(Association for Computational Linguistics, 2020; https://www.
aclweb.org/anthology/2020.nlpcovid19-acl.10
Harder, V. S., Stuart, E. A. & Anthony, J. C. Propensity score
techniques and the assessment of measured covariate balance
to test causal associations in psychological research. Psychol.
Methods 15, 234–249 (2010).
Kaplan, E. L. & Meier, P. Nonparametric estimation
from incomplete observations. J. Am. Stat. Assoc. 53,
457–481 (1958).
Acknowledgements
This research was supported by the US Department of Veterans
Affairs (VA) Office of Research and Development (ORD) Cooperative
Studies Program (CSP) Epidemiology Center at the VA Boston
Healthcare System through CSP #2032, by resources and the use of
facilities at the VA Boston Healthcare System and VA Informatics and
Computing Infrastructure (VINCI) (VA HSR RES 13-457) and by the use
of data from the VA COVID-19 Shared Data Resource. B.A.D., A.L.M.
and M.A.H. were supported by CSP #2032. B.A.D. was supported
by a grant (R00 CA248335) from the National Institutes of Health.
H.G. and B.R.F. were supported by a grant (MVP000) from the VA
Million Veteran Program. M.J.F.M. was supported by a grant (T32
GM140972) from the National Institute of General Medical Sciences
Interdisciplinary Training Program for Biostatisticians. We thank
D. C. Posner, Y.-L. (A.) Ho and K. E. Lynch for insights on COVID-19 data
extraction and phenotype definitions; C. A. Hoag for management of
the administrative and regulatory aspects of the project; J. M. Gaziano
for advice and support; the VA COVID-19 Shared Data Resource team
for their contributions and support; and the VA healthcare providers,
Article
employees and volunteers for their dedication to caring for our
veterans through this pandemic.
Author contributions
B.A.D., H.G. and M.A.H. designed the study. H.G. and K.E.K. designed
the data extraction and assembly workflow. H.G., B.R.F., M.J.F.M.,
N.A. and J.K.W. gathered the data. B.A.D., H.G. and A.L.M. designed
the analytic pipeline. H.G. analysed the data. B.A.D., H.G. and M.A.H.
vouch for the data and analysis. B.A.D. wrote the first draft of the
manuscript. All authors critically reviewed the manuscript and
decided to proceed with publication.
Competing interests
B.R.F. reported completing a 6-month paid internship with Moderna
before joining the Department of Veterans Affairs and owning five
shares of Moderna, Inc. stock purchased in February 2020 outside the
submitted work. M.A.H. reported receiving personal fees from Cytel
and ProPublica outside the submitted work. The other authors declare
no competing interests.
Ethical compliance
This work was approved by the VA Boston Healthcare System
Research & Development Committee and Human Subjects
Subcommittee, and an exemption of informed consent and HIPAA
authorization was granted.
Additional information
Extended data is available for this paper at
https://doi.org/10.1038/s41564-022-01272-z.
https://doi.org/10.1038/s41564-022-01272-z
Supplementary information The online version
contains supplementary material available at
https://doi.org/10.1038/s41564-022-01272-z.
Correspondence and requests for materials should be addressed
to Hanna Gerlovin.
Peer review information Nature Microbiology thanks the anonymous
reviewers for their contribution to the peer review of this work.
Reprints and permissions information is available at
www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Disclaimer: The contents of this article do not represent the views
of the US Department of Veterans Affairs or the US Government.
The views and opinions expressed in this article are those of the
authors and do not necessarily reflect the official policy or position
of the Department of Health and Human Services and its agencies,
including Biomedical Advanced Research and Development
Authority and the Food and Drug Administration, as well as any
other agency of the US Government. Assumptions made within
and interpretations from the analysis do not necessarily reflect the
position of any US Government entity.
This is a U.S. Government work and not under copyright protection in
the US; foreign copyright protection may apply 2023
CAUSALab, Harvard T.H. Chan School of Public Health, Boston, MA, USA. 2Department of Epidemiology, Harvard T.H. Chan School of Public Health,
Boston, MA, USA. 3Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC), VA Boston Healthcare System, Boston, MA, USA.
4
Department of Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA. 5Department of Biostatistics, Boston University
School of Public Health, Boston, MA, USA. 6Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.
7
Division of Aging, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA. 8Department of Biostatistics, Harvard T.H. Chan School of
Public Health, Boston, MA, USA. 9These authors contributed equally: Barbra A. Dickerman, Hanna Gerlovin.
e-mail: hanna.gerlovin@va.gov
1
Nature Microbiology
Article
Extended Data Fig. 1 | Covariate Balance (Love) Plot. Shows the difference
in means for conditions identified by the Centers for Disease Control and
Prevention (CDC) as risk factors for severe COVID-19 in the matched population
from the (a) primary analysis during a period spanning Delta- and Omicron-
Nature Microbiology
https://doi.org/10.1038/s41564-022-01272-z
variant predominance and (b) secondary analysis during a period of Omicronvariant predominance. A strict balance cut-off was set at 0.1. An asterisk denotes
the standardization of continuous variables.
Article
https://doi.org/10.1038/s41564-022-01272-z
Extended Data Fig. 2 | Negative control 1: Cumulative Incidence of Symptomatic COVID-19 in the First 7 Days After the Third Vaccine Dose (October 20, 2021–
February 15, 2022). Solid blue line represents the risk curve for BNT162b2, dashed orange line represents the risk curve for mRNA-1273 and shaded areas represent
pointwise 95% confidence intervals.
Nature Microbiology
Article
https://doi.org/10.1038/s41564-022-01272-z
Extended Data Fig. 3 | Negative control 2: Cumulative Incidence of Non-COVID-19 Death Over the Follow-up (October 20, 2021–February 15, 2022). Solid blue line
represents the risk curve for BNT162b2, dashed orange line represents the risk curve for mRNA-1273 and shaded areas represent pointwise 95% confidence intervals.
Nature Microbiology
Last updated by author(s): October 11, 2022
Reporting Summary
Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist.
nature portfolio | reporting summary
Corresponding author(s): Hanna Gerlovin and Barbra Dickerman
Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.
A description of all covariates tested
A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient)
AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)
For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.
For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Our web collection on statistics for biologists contains articles on many of the points above.
Software and code
Policy information about availability of computer code
Data collection
The electronic health records used in this study come from the health care databases of the Department of Veterans Affairs (VA). The data
was extracted from the VA's Corporate Data Warehouse using Microsoft SQL Server Management Studio version 18.1.
Data analysis
Analyses were performed with R software, version 3.6.0 (R Foundation for Statistical Computing), and SAS Enterprise Guide, version 8.2 (SAS
Institute).
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and
reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information.
Data
- Accession codes, unique identifiers, or web links for publicly available datasets
- A description of any restrictions on data availability
- For clinical datasets or third party data, please ensure that the statement adheres to our policy
March 2021
Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
The data that support the findings of this study are available from the VA. VA data are made freely available to researchers behind the VA firewall with an approved
1
Human research participants
Policy information about studies involving human research participants and Sex and Gender in Research.
Reporting on sex and gender
The primary findings do not apply to only one sex. The analysis matched on sex (as documented in the EHR database) as a
potential confounder for the effect of interest. Subgroup analyses by sex were performed.
Population characteristics
Electronic health records of 130,392 individuals from the U.S. Department of Veterans Affairs (VA) electronic healthcare
databases were analyzed. The median age was 70 years (interquartile range, 62-74 years), 96% of individuals were male, and
24% were Black.
Recruitment
In this study, we identified users of the U.S. Veteran Health Administration (VHA) using U.S. Department of Veterans Affairs
(VA) electronic healthcare databases. The U.S. VHA provides health care to discharged veterans of the U.S. armed forces. It is
the largest integrated healthcare system in the U.S., providing care at 1,293 health care facilities, including 171 VA Medical
Centers and 1,112 outpatient sites of care located across the U.S. Our specific study population was comprised of regular VA
users – who were predominantly men and older persons – which may limit the generalizability of our findings.
Ethics oversight
This study was approved by the VA Boston Healthcare System Research & Development Committee and Human Subjects
Subcommittee.
nature portfolio | reporting summary
VA study protocol. More information is available at https://www.virec.research.va.gov or by contacting the VA Information Resource Center (VIReC) at
VIReC@va.gov.
Note that full information on the approval of the study protocol must also be provided in the manuscript.
Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences
Behavioural & social sciences
Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf
Life sciences study design
All studies must disclose on these points even when the disclosure is negative.
Sample size
We used VA electronic health records for 130,392 matched individuals, as detailed in the Methods. b. The sample size was determined via
application of the eligibility criteria and matching algorithm, as detailed in the manuscript.
Data exclusions
The 130,392 matched individuals included satisfied the pre-specified eligibility criteria, as detailed in the Methods.
Replication
Replication was not applicable to this study, which aimed to evaluate the comparative effectiveness of third doses of mRNA-based Covid-19
vaccines using the information on all eligible individuals in the database.
Randomization
No randomization was involved in our analysis of observational electronic health record data.
Blinding
Investigator blinding was not relevant to our analysis of observational electronic health record data.
Reporting for specific materials, systems and methods
We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.
Methods
n/a Involved in the study
n/a Involved in the study
Antibodies
ChIP-seq
Eukaryotic cell lines
Flow cytometry
Palaeontology and archaeology
MRI-based neuroimaging
March 2021
Materials & experimental systems
Animals and other organisms
Clinical data
Dual use research of concern
2